Developing ‘travelthai’: a dynamic Computable general equilibrium model for tourism of Thailand and case applications on tourism setbacks and tourism-related fiscal policies

This thesis is aimed to develop a Monash-style medium-scale dynamic computable general equilibrium (CGE) model for Tourism of Thailand as the major contribution. The model is designed to be capable of analyzing not only the economic policies but also historical simulation and forecasting simulation. The comparative static version of the ‘THORANI’ model is first developed and later on improved it into the dynamic version of the ‘TRAVELTHAI’ model that comprises dynamic capability. TRAVELTHAI’s modeling structure is based upon the MyAGE and McHuge, which are successors of MONASH, the dynamic CGE model of the Australian economy (Dixon and Rimmer 2002) that contains the dynamic mechanism features. The TRAVELTHAI model captures the structure of the Thai economy using the coefficients attributing from the Thai Tourism Satellite Accounts (TSAs) which is the main distinction from the other kinds of CGE models, for which normally use the Input Output Tables, the adoption of proper substitution elasticities as the key parameters, and the portrayal of specific features regarding tourism which are also relevant to the Thai economy. The model contains a detailed disaggregation of industries, especially those related to Tourism, factors of production, technologies and preferences, taxes and margins, a comprehensive illustration of government accounts, balance of payment accounts, and net foreign liabilities. The dynamic mechanisms also include the links between stock and flow variables for physical capital accumulation, for debt and savings, and for wage-employment adjustment. TRAVELTHAI possesses 3 types of tourism demands: Inbound, Domestic, and Outbound tourism. Although the thesis thematically focuses on inbound tourism or travel receipts in the Service accounts of the Balance of Payments, this gives the power to further study in terms of revenue and expenditure of the balance of travel. The baseline historical simulation of the Thai economy over the period 2001-2011 and baseline forecast for 2012-2020 are performed. The decade has shown an interesting rapid growth with the contribution of tourism growth of seven percent per annum on the average despite adverse tourism setbacks ranging from epidemic flu, tsunami, the recent political turmoil and the flood. Six major simulations with TRAVELTHAI have been conducted. The first is the historical simulation, for which observed changes in economic variables e.g. changes in GDP, outputs of industries, factors of production, household consumption or international trade become the exogenous shocks for the model, and changes in technologies and preferences are solved endogenously. This is aimed to develop the baseline for the Thai economy. Three projections to the year 2020 of the forecasting simulation have been carried out, bases on the evidence of the decade average, the first half and the second half of the previous decade. This time, the economic performance is evaluated from the structural changes in technologies and preferences obtained earlier and projects to the future to develop the baseline forecasts of these three projections. The fifth and the sixth are major simulations regarding the thesis objectives. They are those with external shocks and policy simulations. Simulation for external shocks of tourism setback in term of recent major flood is delivered. Behavioral effects of demand side and resource losses effect of supply side shocks are then applied in five sub-total shocks. Two behavioral effects are inbound tourism shift and the cut in the number of domestic tourists. Other three resources losses shocks are related to the losses or damages of factors of production: labor, capital, and land. The sixth simulation is the tourism-related fiscal policy on the tax cut for inbound tourism. The comparison between tax cut and without tax cut scenarios is also presented. This is aimed to pursue for alleviating measure to soothe the effect of tourism setback. Apart from historical and forecasting simulations which are aimed to create the benchmark as the baseline forecasts, the results from the setback simulation show that in the short run the flood causes the declines of real GDP and its compositions relative to the baseline. Trade balance deteriorates. Inbound tourism also decreases but not as much as outbound and domestic tourism. Domestic tourism revenue, thus, drops. Travel receipts accordingly weaken, but less than travel expenses worsen, which improves travel balance. For the supply side, the labors migrate to non-tourism industries from the tourism-related industries. Output productions are all reduced, with non-tourism industries most worsening. The flood damages the economy. In all, resource losses effects show domination over other effects. The after effect forecast growths to the year 2020 show that both behavioral and resource-loss effects of flooding reduce real GDP and its components overtime. Labors migrate to the non-tourism industries from direct tourism industries and tourism-connected industries. Direct tourism industries suffer more than tourism-connected industries in terms of labor reduction and the releases of labor rising through the period. All sectors produce less output, accordingly. Over-time slight deterioration in terms of trade subsequently improves balance of travel. Simulation on a tax cut as an incentive for the inbound tourist shows a plus to the falling real GDP relative to the baseline. Tax cut creates a larger impact on inbound tourism. Travel receipts increase accordingly, comparing to the decrease in the scenario without any tax cut. Tax cut improves balance of travel towards a surplus comparing to the lower gain of that without the tax cut. It also generates more outputs in the direct tourism industries as expected, though not enough to make a positive outcome for the total but the fall in the reduction of output production is a better off. This is in line with the greater amount of labor hired in the sector. For other two industries, they are worse off as their own labors migrate to the direct tourism industries. Their productions follow the falls in their sectoral employment. For the after effect tax cut forecast growth to 2020, only real inbound tourism and tourism receipts have considerable changes comparing to that without tax cut. For real inbound tourism, the after-flood growth is negative for without tax cut scenario in the short run and recover to be positive in the long run. With the tax cut, the deviation from base case forecast is positive and growing further in the long run. Tourism receipts show similar pattern. Nonetheless, the effect on the deviation over real GDP is very tiny and it reveals the same trend for both tax cut and without tax cut. Outputs and employments of direct tourism industries with tax cut decline less than that of without tax cut, though with similar declining trend. For tourism-connected industries, the reductions are about the same with the similar trend. For non-tourism industries, there are further output and employment reductions comparing to the without tax cut scenario. Tax cut incentive could be one of alleviating measures to soothe the tourism setback. Political-wise, a mix of incentives could be further pursued for what-if analyses to examine all inclusive effects other than tourism-focused. Such policy planning is helpful to understand trade-offs among economic agents to minimize incurring costs.